614
There are also a variety of methods of
representing features. Point and linear
features are most commonly represented by
one or more points; with implicit rules for
joining the points to make lines. This is
known as a vector technique.
The most common type of line is defined
by joining consecutive points in straight
line segments. The variances in vector
representations tend to centre around the
means of storing their interrelation.
There are several ways in which this can
be done. Points may be stored in tables
and identified by features. A series of
points may be stored as a separate entity.
Either case requires another series of tables
or pointers to determine the
interrelationship of features.
While polygonal features are often
similarly represented, polygonal covers
tend to get separate attention due to then-
prevalence and complexity. There are
three commonly used techniques. The
most intuitive technique represents a
polygon’s position by identifying the lines
that form its boundaries. A similar
technique, more commonly used in
analysis, stores the polygon’s position by
associating the polygons to the left and
right with each boundary line. One of the
surface models is applicable, as a
polygonal cover has properties in common
with a surface. The raster technique
provides a pointer to the containing
polygon for each pixel.
TINs require numerous positional
coordinates and a series of pointers or
tables in order to operate efficiently.
There are many ways in which this can be
done. Uncompressed tessellation models
require large quantities of data. Common
means of compressing this type of data are
run-length encoding or quad-trees.
While numerous data formats can be used,
most of these formats either consist of, or
can easily be translated into, some
combination of:
• points or lines represented as a
series of points,
• matrices of values,
• non-graphic attributes,
• and interrelationships between the
above.
Data Formats
For each model, there are many storage
formats. Most vector techniques including
